The idea, in plain English
Start here.
Keyword research means finding and evaluating the words people use to look for information, services or products. Start with real questions, expand your ideas and decide which tasks your website can answer well. Tool estimates help you compare opportunities; they do not promise visitors or sales.
A spreadsheet with thousands of phrases can look impressive while leaving the next decision unclear. In this lesson, you will build a smaller research sheet that explains where each idea came from, what someone probably needs and whether you have a useful answer. You can begin without buying a tool or inventing missing metrics.
A few words you’ll meet.
OPEN WHAT YOU NEED ↙Seed topic
A starting subject for research, such as faucet repair. It is a prompt for investigation, rather than a final page assignment.
Search volume
A tool’s estimate of how often a query is searched in a stated market and period. Searches are not the same as people, visits or sales.
Keyword difficulty
A provider’s model of ranking difficulty. Methods differ, so it is not a Google score or a universal pass mark.
Keyword cluster
A group of related queries that may be served by the same useful page. Shared words alone do not prove shared intent.
Listen before opening a keyword tool
Write down the services, products or problems your website actually covers. Then collect the language people use: questions from support, sales conversations, site search or approved research notes. Remove personal information. The goal is to learn how a person describes the need, rather than translate every question into marketing language.
For the fictional Harbor Workshop, ‘Why does the tap drip after I close it?’ is a useful starting point. It suggests a real question that a repair guide might explain. The separate request ‘Can I book a planned repair?’ suggests a service task. Keep both as research ideas until you have checked the evidence. A list of customer questions is a good starting artifact.
Expand a topic without mistaking ideas for demand
Use a few seed topics to explore related questions and wording. A keyword tool, public search suggestions and relevant competitor pages can introduce ideas you had not considered. Open promising results and note the information offered. Keep the source and date so another person can understand where the idea came from.
AI can help brainstorm possible phrases, but a plausible phrase is not a measured search query. Label it as an idea until you check it. If you already have access to the site’s Search Console, its performance reports offer evidence about searches in which your site appeared. That is evidence about your site’s recorded visibility, rather than a complete measurement of everyone’s demand.
- Keep customer wording alongside any shorter keyword version.
- Separate brainstorming, tool estimates and observed site data.
- Record the market, date and tool when adding a metric.
From an idea to a useful page decision
Use metrics to ask better questions
An estimated monthly search volume can help compare topics, but it does not tell you how many visits your page will receive. Some searches repeat, search results can satisfy a need without a visit, and your page may appear for several related queries. A low or missing estimate is a reason to investigate relevance, not automatic proof that a question is worthless.
Difficulty scores also need context. Ahrefs describes its own score using linking websites to top-ranking pages; another provider may use a different method. Do not combine scores as if they were one scale or choose a universal cutoff. Inspect the actual competing answers and consider what your team can credibly contribute. Advertising cost estimates describe a paid-search market, not guaranteed organic customer value.
Shortlist by fit, then prepare a page decision
Ask three practical questions: Does this need matter to our audience? Can we provide an accurate, useful answer? Is there an existing page that should own the answer? These questions keep a popular but unrelated topic from displacing a relevant customer problem.
Group close variations only when they ask for substantially the same answer. ‘Why does a faucet drip?’ and ‘Causes of a dripping faucet’ may belong together; a repair booking request has a different primary task. Your shortlist is an editorial judgment supported by evidence, rather than a prediction of a particular ranking. The next lesson turns those groups into a page map.
- Keep: relevant need, useful contribution and a plausible page role.
- Investigate: uncertain intent, weak evidence or unclear fit.
- Set aside: unrelated demand or a question you cannot responsibly answer.
A shortlist with useful uncertainty
Illustrative research notes for Harbor Workshop. These queries, judgments and missing estimates are fictional teaching examples; no search volumes or rankings were measured.
| Query idea | Evidence available | Decision |
|---|---|---|
| Why does my faucet keep dripping? | A recurring customer question in this example; volume not checked | Investigate a troubleshooting guide. |
| Schedule a faucet repair | Fits the business’s actual planned-repair offer | Check the existing service page. |
| Luxury bathroom inspiration | Related subject, but outside this example’s offer | Set aside for this project. |
| Faucet repair cost | Useful customer question; actual scope and prices needed | Research what the business can accurately explain. |
Put it into practice
Build a six-query shortlist
Choose a familiar fictional business, a personal project or a site you are authorized to research. You do not need paid keyword software.
- Collect six query ideas and record where each came from.
- Describe each likely task in one sentence.
- Add any real metric with its tool, market and date; otherwise write ‘unknown’.
- Mark each idea keep, investigate or set aside, with one reason.
- Choose two ideas and check whether they need one page or different answers.
Keep your working notes in your own document. This page saves only your completion check, not your answers or customer data.
Check your understanding
Make the call.
Choose an answer for each question. Then compare the reasoning.
You have a useful next step.
- Start with real questions and preserve their source.
- Treat tool metrics as contextual estimates and leave unknowns visible.
- Choose useful page opportunities by audience fit, evidence and existing ownership.
Completion is your own learning record. Quiz scores and practice checks are saved separately, on this browser.
Sources & context
The sources below explain the factual foundations and editorial methods used in this lesson. Worked examples, planning models and exercises are illustrative teaching material.
- Google SEO Starter Guide
Official guidance on useful content and the words readers may use.
- Google: Performance report
Primary definitions and scope of Search Console search-performance data.
- Ahrefs: Keyword research guide
Original provider explanation of its own difficulty methodology and research metrics; no ranking promises adopted.